Human Specific Vitamin C Metabolism And Xenobiotic Polymorphism: The Optimal Nutrition

The biomedical significance of human vitamin C (VC) metabolism is reviewed in the light of polymorphisms in xenobiotic enzymes deduced from genetic, biochemical, and epidemiological results to estimate optimal nutrition. VC comprises both ascorbic acid (AsA) and dehydroascorbic acid (DAsA). AsA is oxidized to DAsA via short-lived monodehydroascorbate radicals in a series of xenobiotic reactions and by reactive oxygen species (ROS). DAsA is reversibly reduced by glutaredoxin, but is also irreversibly hydrolyzed into 2,3-diketo-L-gulonate by dehydroascorbatase [EC 3.1.1.17] and non-enzymatic reactions. VC is a cofactor in reactions catalyzed by Cu+-dependent monooxygenases [EC 1.13.12.-] and Fe2+-dependent dioxygenases [EC 1.13.11.-]. VC plays a protective role against oxidative stress by ROS and xenobiotics, via monodehydroascorbate radicals. The Vitamin Society of Japan has re-evaluated old data because of the development of life science. The recommended dietary allowance (RDA) of VC is 100 mg/day for adults in Japan to prevent scurvy. RDA is defined as EAR+2SD, i.e. estimated average requirement (EAR) and the standard deviation (SD) obtained by short-term depletion-repletion studies. However, based on VC synthetic rates in rat, Pauling proposed that the optimum intake is 2.3 g/day. This is the problem of RDA vs. optimal nutrition. Optimal nutrition is wider in scope than RDA that covers genetic polymorphisms, long-term health outcome during the lifespan, and xenobiotics. Humans (VC auxotrophs) have relatively low plasma AsA levels and high serum uric acid levels compared to most VC-synthesizing mammals (VC autotrophs) due to gene defects in L-gulonolactone oxidase (GLO [EC 1.1.3.8]) and uricase (urate oxidase) [EC 1.7.3.3], respectively. Extrapolation of metabolic data of VC autotrophs to estimate human optimal nutrition is limited because of the compensatory mechanism for the GLO defect in VC auxotrophs, including DAsA transport by GLUT1, and specific mutations in uricase and dehydroascorbatase. Beneficial effects of long-term VC supplementation remain controversial, perhaps because of 1. genetic heterogeneity in study populations, and 2. the balance of antioxidant and pro-oxidant activities of VC depending on the xenobiotic conditions. Thus, in addition to the biochemical studies on AsA and DAsA, human genetic analysis on VC-loading experiments and epidemiological survey are needed. There are marked interindividual differences (coefficient of variation >45%) in the metabolism of VC. This difference is evident during oral loading with 1 mmol AsA or DAsA in subjects consuming a diet low in VC (less than 5 mg/day) for 3 days before loading in the cross-over experiment. Since tubular maximum reabsorption of AsA (TmAsA) and glomerular filtration rate (GFR) are similar among subjects, degradation steps of VC may be involved in the personal difference. The metabolisms of three most important water-soluble antioxidants in mammals i.e. VC, urate and glutathione are different in humans and other animals. The effects of polymorphism A313G (I1e105Val) in the gene for glutathione S-transferase P1 (GSTP1) [EC 2.5.1.18], one of the most active xenobiotic enzymes in the second phase of detoxification, on human VC metabolism were thus studied. In an epidemiologic survey of Mongolians (n = 164) with very low VC intake, serum VC concentration was only 28%, and the level of reactive oxygen metabolites was 128%, when compared with those in Japanese. The variant frequency of GSTP1 among Japanese subjects (n = 210) was AA, 71.0%; GA 27.0% and GG, 1.9%. In Mongolian subjects (n = 93), it was AA, 62.4%; GA, 36.6%; and GG, 1.1%. In VC loading experiments, at 24 h after administration of 1 mmol of VC to young women (n = 17; age, 21.0 +/- 1.1 y, glomerular filtration rate, GFR = 90 ml/min), total VC excretion (46.7 +/- 18.1 mg) by AA homozygotes of GSTP1 was greater (p < 0.0069) than that (28.2 +/- 14.0 mg) by GA heterozygotes. One hour after administration of VC, blood total VC levels were also significantly different (p < 0.0036) between the homozygotes and heterozygotes. The results of background experiments were as follows: (1) the VC level in 24-h urine after VC loading did not differ between the two orally administered C forms (AsA and DAsA); (2) VC excretion between 0 and 3 h after VC loading was significantly higher (p < 0.05) for DAsA, while those between 3 and 6, 6 and 9, 9 and 12, and 12 and 24 h after VC loading were significantly higher (p < 0.05 or p < 0.01) for AsA; and (3) blood VC concentrations and the increase in VC at I h after VC loading were significantly higher (p < 0.05 and p < 0.01, respectively) in the DAsA group than in the AsA group. The difference between AsA and DAsA dynamics in (2) and (3) may be explained by the sodium-dependent active transport of AsA by SVCT1 and 2, and passive transport of DAsA by glucose transporters (GLUTs) in the presence of glutathione. The large species differences in DAsA metabolism are partly explained by the low activity of human dehydroascorbatase, which has a unique structure, as deduced by X-ray crystallography, and a unique sequence of 299 amino acids. The anti-oxidant and anti-xenobiotic roles of monodehydroascorbate radicals both in vivo and in vitro are important. ROS are generated mainly in mitochondria but DAsA transported through GLUT1 into mitochondria is converted into AsA and prevents oxidative stress. Finally RDA and optimal nutrition are discussed from the standpoint of human specific metabolism of VC including prevention against ROS produced by exercise and pathological conditions.